Review
Hamstring strength and exibility after hamstring
Additional material is
published online only. To view
please visit the journal online
(http://dx.doi.org/10.1136/
bjsports-2015-095311).
1 Design Systematic review and meta-analysis.
School of Exercise Sciences,
Australian Catholic University,
Melbourne, Victoria, Australia
2
School of Exercise and
Nutrition Sciences and Institute
of Health and Biomedical
Innovation, Queensland
University of Technology,
Brisbane, Queensland,
Australia
3
School of Health, Sport and
Professional Practice, University
of South Wales, Pontypridd,
Wales, UK
Correspondence to
Nirav Maniar, School of
Exercise Sciences, Australian
Catholic University, 17 Young
Street, Fitzroy 3065, VIC,
Australia; Nirav.Maniar@acu.
edu.au
Accepted 19 March 2016
Published Online First
13 April 2016
To cite: Maniar N,
Shield AJ, Williams MD,
et al. Br J Sports Med
2016;50:909920.
strain injury: a systematic review and meta-analysis
Nirav Maniar,1 Anthony J Shield,2 Morgan D Williams,3 Ryan G Timmins,1
David A Opar1
ABSTRACT
Objective To systematically review the evidence base
related to hamstring strength and exibility in previously
injured hamstrings.
Data sources A systematic literature search was
conducted of PubMed, CINAHL, SPORTDiscus, Cochrane
Library, Web of Science and EMBASE from inception to
August 2015.
Inclusion criteria Full-text English articles which
included studies which assessed at least one measure of
hamstring strength or exibility in men and women with
prior hamstring strain injury within 24 months of the
testing date.
Results Twenty-eight studies were included in the
review. Previously injured legs demonstrated decits
across several variables. Lower isometric strength was
found <7 days postinjury (d=1.72), but this did not
persist beyond 7 days after injury. The passive straight
leg raise was restricted at multiple time points after
injury (<10 days, d=1.12; 1020 days, d=0.74;
2030 days, d=0.40), but not after 4050 days
postinjury. Decits remained after return to play in
isokinetically measured concentric (60/s, d=0.33) and
Nordic eccentric knee exor strength (d=0.39). The
conventional hamstring to quadricep strength ratios
were also reduced well after return to play (60:60/s,
d=0.32; 240:240/s, d=0.43) and functional
(30:240/s, d=0.88), but these effects were
inconsistent across measurement methods.
Conclusions After hamstring strain, acute isometric
and passive straight leg raise decits resolve within 20
50 days. Decits in eccentric and concentric strength and
strength ratios persist after return to play, but this effect
was inconsistent across measurement methods. Flexibility
and isometric strength should be monitored throughout
rehabilitation, but dynamic strength should be assessed
at and following return to play.
INTRODUCTION
Hamstring strain injuries (HSIs) are the most common
non-contact injuries in Australian rules football,15
soccer,610 rugby union,1114 track and eld1517 and
American football.18 HSIs result in time away from
competition,9 nancial burden9 19 and impaired per-
formance on return to competition.20
Further to this, recurrent HSI often leads to a
greater severity of injury than the initial insult.10 14
The most commonly cited risk factor for future HSI
is a previous HSI.2124 The high recurrence rates of
HSI10 14 are proposed to result from incomplete
recovery and/or inadequate rehabilitation25 26
because of pressure for early return to play at the
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
expense of convalescence.27 Consequently, there has
been much interest recently in observations of ham-
string structure and function in previously injured
legs compared with control data.2834 Despite the
possible limitation of this approach, it is often
agreed that decits that exist in previously injured
hamstrings could be a maladaptive response to
injury.35 As such, these decits that persist beyond
return to play could provide markers to better
monitor athletes during and/or at the completion
of rehabilitation.35
Which parameters are the best markers to monitor
an athletes progress during rehabilitation?
Conventional clinical practice focuses on measures
of strength and exibility; however, the evidence is
based on predominantly retrospective observations
of strength,28 29 3642 strength ratios36 37 39 40 43 44
and exibility26 28 42 4549 in previously injured ath-
letes. These studies are limited in reporting single or
isolated measures with methodologies and popula-
tions that differ from study to study. To advance
knowledge, we aimed to systematically review the
evidence base related to hamstring strength and
exibility in previously injured hamstrings.
METHODS
Literature search
A systematic literature search was conducted of
PubMed, CINAHL, SPORTDiscus, Cochrane
Library, Web of Science and EMBASE from incep-
tion to August 2015. Keywords (table 1) were chosen
in accordance with the aims of the research.
Retrieved references were imported into EndNote X7
(Thomson Reuters, New York City, New York, USA),
with duplicates subsequently deleted. To ensure all
recent and relevant references were retrieved, citation
tracking was performed via Google Scholar and refer-
ence list searches were also conducted.
Selection criteria
Selection criteria were developed prior to searching to
maintain objectivity when identifying studies for inclu-
sion. To address the aims, included papers had to:
Assess at least one parameter of hamstring
strength (maximum strength, associated strength
ratios and angle of peak torque) or exibility in
humans with a prior HSI within 24 months
from the time of testing;
Have control data for comparison (whether it
was a contralateral uninjured leg or an uninjured
group);
Have the full-text journal article in English
available (excluding reviews, conference
abstracts, case studies/series);
1 of 14
 Review
Table 1 Summary of keyword grouping employed during database
searches
Muscle group Injury
Hamstring* Injur*
Semitendinosus Strain*
Semimembranosus Tear
Biceps Femoris Rupture*
Posterior Thigh Pull*
Thigh Trauma
Torn
*Truncation. Boolean term OR was used within categories, while AND was used
between categories.
Not include hamstring tendon or avulsion injuries.
The titles and abstracts of each article were scanned by one
author (NM) and removed if information was clearly inappropriate.
Selection criteria were then independently applied to the remaining
articles by three authors (NM, RGT and DAO). Full text was
obtained for remaining articles, with selection criteria reapplied by
one author (NM) and cross-referenced by another author (DAO).
Analysis
Assessing bias and methodological quality
Risk-of-bias assessment was performed independently by two
examiners. We used a modied version of a checklist by Downs
and Black.50 The original checklist contained 27 items, however
many were relevant only to intervention studies. Since the
majority of the papers in this review were of a retrospective
nature, items 4, 8, 9 13, 14, 15, 17, 19, 22, 23, 24 and 26 were
excluded as they were not relevant to the aims of the review.
Of the remaining items, 1, 2, 3, 5, 6, 7 and 10 assessed
factors regarding the reporting of aims, methods, data and
results, while items 16, 18, 20, 21 and 25 assessed internal val-
idity and bias. Item 27 was not suitable to the context of the
current review and was modied to address power calculations.
Two new items (items 28 and 29) relating to injury diagnosis
and rehabilitation/interventions were added to more appropri-
ately assess the risk of bias; thus, the modied checklist con-
tained 17 items (see online supplementary table S1).
Fourteen of the items were scored 0 if the criterion was not
met or it was unable to be determined, while successfully met
criteria were scored 1 point. The other three items (items 5, 28
and 29) were scored 0, 1 or 2 points, as dictated by the criteria
presented in online supplementary table S1. This resulted in a
total of 20 points available for each article.
Similarly, modied versions of this checklist have been used in
previous systematic reviews investigating factors leading to heel
pain51 and risk factors associated with hamstring injury.52 The
risk-of-bias assessment was conducted by two authors (NM and
DAO), with results expressed as a percentage. In the case of dis-
agreement between assessors, an independent individual was
consulted with consensus reached via discussion if necessary. In
situations where one of the assessors (DAO) was a listed author
on a study included for review, the independent individual com-
pleted the risk-of-bias assessment in their place.
Data extraction
Relevant data were extracted including the participant numbers,
population and sampling details, diagnosis technique, severity of
injury, time from injury to testing (in days assuming 30.4 days/
month, 365 days/year), variables investigated and how these were
tested, results including statistical analysis and, where appropriate,
2 of 14
potential confounders that may affect strength or exibility out-
comes. The major confounders include other lower limb injuries
likely to affect strength and exibility, interventions and rehabili-
Time tation programmes performed. Furthermore, insufcient evidence
exists regarding the interaction between gender and HSI; thus,
Past
mixed-gender cohorts were considered as a potential confounder.
Prior
Retrospective*
Data analysis
Previous*
Although objectively synthesising evidence via a meta-analysis is
Recent*
often desirable, this technique was not able to be applied to all
Histor*
the evidence retrieved in this review owing to insufcient
reporting of data (ie, two or more studies or subgroups with
mean, SD and participant numbers for contralateral leg compar-
isons) or methodological variations between studies.
When sufcient data were available, meta-analysis and graph-
ical outputs were performed using selected packages5355 on R
(R Development Core Team. R: A language and environment
for statistical computing. Vienna, Austria: R Foundation for
Statistical Computing, 2010). Standardised mean differences
(Cohens d) facilitated the comparison of studies reporting vari-
ables in different units, with effect estimates and 95% CIs sum-
marised in forest plots. A random-effects model was used to
determine the overall effect estimate of all studies within the vari-
able or subgroup as appropriate, with variance estimated through
a restricted maximum likelihood (REML) method. The magni-
tudes of the effect size were interpreted as small (d=0.20), moder-
ate (d=0.50) and large (d=0.80), according to thresholds
proposed by Cohen.56 Where studies reported multiple types of
data (eg, multiple isokinetic velocities, multiple subgroups or mul-
tiple time points), these data were analysed as subgroups to avoid
biasing the weighting of the data. These time points were dictated
by the data available. Where data were available in the acute stages
(prior to return to play), time bands were kept at <10 days as it
would be expected that decits would change relatively rapidly
during this time owing to ongoing rehabilitation and recovery.
Data presented for participants at or after return to play were
pooled for two reasons: (1) no included study reported any
ongoing rehabilitation after return to play and (2) many of these
studies had variable time from injury until testing between in-
dividual participants. Where a study had multiple time points
that t within post return to play time band (eg, at return to
play and follow-up), the earlier option was chosen as there was
expected to be a lower chance of bias due to other uncontrolled
or unmonitored activities. For the purposes of meta-regression
(employed to assess the effects of time since injury), studies with
multiple time points were pooled to provide the best assessment
of the effect of time on the given variable. By considering each
time point as a subgroup, sufcient data (>10 subgroups) were
available for meta-regression analysis57 providing that time
from injury until testing was reported. Funnel plots were visu-
ally inspected for asymmetry to assess publication bias.
Heterogeneity was determined by the I2 statistic and can be
interpreted via the following thresholds:57
040%: might not be important
3060%: may represent moderate heterogeneity
5090%: may represent substantial heterogeneity
75100%: considerable heterogeneity.
In situations where it was deemed that reported data (ie, mean,
SD, participant numbers for contralateral leg comparisons) were
insufcient for meta-analysis and could not be obtained via
supplementary material or from contacting the corresponding
author, a best evidence synthesis58 was employed. The level of
evidence was ranked according to criteria consistent with previ-
ously published systematic reviews59 60 as outlined below:
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311

Strong: two or more studies of a high quality and generally con-
sistent ndings (75% of studies showing consistent results);
Moderate: one high-quality study and/or two or more low-
quality studies and generally consistent ndings (75% of
studies showing consistent results);
Limited: one low-quality study;
Conicting: inconsistent ndings (<75% of studies showing
consistent results);
None: no supportive ndings in the literature.
A high-quality study was dened as a risk-of-bias assessment
score of 70%, whereas a low-quality study had a risk of bias
assessment score of <70%.57
Figure 1 Flow diagram outlining
steps for study inclusion/exclusion.
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
Review
RESULTS
Search results
The search strategy consisted of six steps (gure 1). The initial
search yielded 7805 items (Cochrane Library=131;
PubMed=2407, CINAHL=604; SPORTDiscus=640; Web of
Science=1049; EMBASE=2974) from all databases. After
duplicates were removed, 4306 items remained. Title and
abstract screening resulted in 92 remaining articles, reference list
hand searching and citation tracking resulted in the addition of
six articles. Independent application of the selection criteria
yielded 28 articles to be included in the review, 22 of which
were included in meta-analysis.
3 of 14
 Review

Risk-of-bias assessment Outcomes

Risk-of-bias assessment of each article is displayed in table 2. It
is important to note that the risk-of-bias assessment was not the
basis of exclusion. Included articles ranged from a score of 8 to
18 of a possible 20 (4090%).
Description of studies
Participants
A sample of 898 participants (n=802 male, n=96 female; age
range, 1547 years) were examined across the included studies. A
total of 17 studies included only male participants,29 34 36 37 3943
45 46 48 49 6164
10 studies had mixed gender,26 28 33 47 6570
while only 1 exclusively studied females.71 Participants were gen-
erally considered recreationally active at a minimum.
Injury
Methods of diagnosis varied between studies, with some studies
using multiple methods of diagnosis. A total of 12 studies used
clinical criteria,26 28 33 34 36 37 42 48 6669 10 used
MRI,26 28 29 33 34 62 65 6769 5 had medical or health practi-
tioner diagnosis,39 41 43 48 7 used a questionnaire or self-
report,40 46 47 49 58 63 71 2 used ultrasound36 37 and 2 had
unclear methods of diagnosis.45 70 Description of severity of
injury varied signicantly between studies, with the most
common being time to return to play26 28 29 4043 48 49 63 67
and grade (IIII) of injury.29 31 33 39 62 66 68 69 70 Description of
time from injury to testing varied signicantly between studies
(range, 2690 days).
The strength variables examined were concentric, eccentric and
isometric strength (absolute and normalised to body mass),
strength ratios (usually hamstring to quadricep (H:Q)) and angle
of peak torque. The ve exibility variables examined were
passive straight leg raise, active straight leg raise, passive knee
extension, active knee extension and the sit and reach. All ve
strength variables (concentric, eccentric, isometric, strength
ratios and angle of peak torque) and three exibility variables
(passive straight leg raise, active knee extension and passive
knee extension) were included for meta-analysis. Sufcient data
were available to run meta-regression analysis for isometric
strength and the passive straight leg raise. The best evidence syn-
thesis method was applied to remaining variables for which
insufcient data were available for meta-analysis. The best evi-
dence synthesis is summarised in table 3.
Strength
Concentric strength
Data for all studies which examined concentric strength can be
found in online supplementary table S2.
Meta-analysis. Concentric strength was measured isokinetically
at 60,29 40 48 6163 66 67 71 18029 40 61 71 and 300/s.39 40 62 71
A statistically signicant small effect for lower concentric strength
at 60/s was found in previously injured legs (effect size 0.33;
95% CI 0.53 to 0.13; I2 0%), but no signicant effects were
found at 180 or 300/s (gure 2).

Table 2 Itemised scoring of study quality using a modified (see online supplementary table S1) Downs and Black checklist (50)
First author, year 1 2 3 5 6 7 10 11 12 16 18 20 21 25 27 28 29 Total Per cent Quality

Arumugam, 2015 0 1 1 1 1 1 1 0 0 1 1 1 1 0 0 0 0 10 50 Low

Askling, 2006 1 1 1 1 1 1 0 0 0 1 1 1 1 1 0 2 1 14 70 High
Askling, 2010 1 1 1 1 1 1 0 0 0 1 1 1 1 0 0 2 0 12 60 Low
Brockett, 2004 1 1 1 1 1 0 1 0 0 1 1 1 1 0 0 1 1 12 60 Low
Croisier, 2000 1 1 1 1 1 1 0 0 0 1 1 1 0 1 0 1 0 11 55 Low
Croisier, 2002 1 1 1 1 1 1 0 0 0 1 0 1 1 1 0 2 0 12 60 Low
Dauty, 2003 1 1 0 1 1 1 0 0 0 1 1 1 1 1 0 1 0 11 55 Low
Doherty, 2012 1 1 1 1 0 1 1 0 0 1 1 1 1 1 0 0 0 11 55 Low
Hennessey, 1993 1 1 0 0 1 1 0 0 0 1 1 1 1 0 0 0 0 8 40 Low
Jonhagen, 1994 1 1 1 1 1 1 1 0 0 1 1 1 1 1 0 0 0 12 60 Low
Lee, 2009 1 1 1 1 1 1 1 0 0 1 1 0 1 1 0 1 0 12 60 Low
Lowther, 2012 1 1 1 1 1 1 1 0 0 1 1 1 1 0 0 0 0 11 55 Low
Mackey, 2010 1 1 1 1 1 1 1 0 0 1 1 1 1 0 0 0 0 11 55 Low
Opar, 201341 1 1 0 1 1 1 1 0 0 1 1 1 1 1 0 2 0 13 65 Low
Opar, 201329 1 1 1 2 1 1 1 0 0 1 1 1 1 1 0 1 1 15 75 High
Opar, 2015 1 1 1 2 1 1 1 0 0 1 1 1 1 1 1 2 0 16 80 High
OSullivan, 2009 1 1 1 1 1 1 1 0 0 1 1 1 1 1 0 0 0 12 60 Low
OSullivan, 2009 1 1 1 1 1 1 1 0 0 1 1 1 1 1 0 0 0 12 60 Low
OSullivan, 2008 1 1 1 1 1 1 0 0 0 1 1 1 1 1 0 0 0 11 55 Low
Reurink, 2015 1 1 1 2 1 1 1 0 0 1 1 1 1 1 1 2 2 18 90 High
Reurink, 2013 1 1 1 2 1 1 1 0 0 1 1 1 1 1 1 2 2 18 90 High
Sanfilippo, 2013 1 1 1 2 1 1 1 0 0 1 1 1 1 1 0 2 2 17 85 High
Silder, 2010 1 1 1 1 1 1 1 0 0 1 1 1 0 1 0 2 1 14 70 High
Silder, 2013 1 1 1 2 1 1 1 0 0 1 1 1 1 1 0 2 2 17 85 High
Sole, 2011 1 1 1 1 1 1 1 0 0 1 1 1 0 1 0 1 0 12 60 Low
Timmins, 2015 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 2 1 16 80 High
Tol, 2014 1 1 0 1 1 1 1 0 0 1 1 1 1 1 0 2 2 15 75 High
Worrell, 1991 1 1 1 1 1 1 0 0 0 1 1 1 1 1 0 0 0 11 55 Low
A high-quality study was defined as a risk-of-bias assessment score of 70%, whereas a low-quality study had a risk-of-bias assessment score of <70%.

4 of 14 Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311

 Review

Table 3 Best evidence synthesis data for all major categories of outcome variables assessed in individuals with a prior hamstring strain injury
not included in the meta-analysis.
Consistency (%)

No. of No Quality Level of evidence

Variable Testing method studies Decrease change Increase (meanSD) of difference

Concentric strength* Seated isokinetic (240/s) 3 100 0 0 6716 Moderate

Eccentric strength Seated isokinetic (30/s) 4 75 25 0 583 Moderate
Eccentric strength Seated isokinetic (120/s) 2 100 0 0 584 Moderate
Isometric Strength Hip 0; knee 90 1 100 0 0 90 Moderate
Concentric strength Seated isokinetic (270/s) 1 100 0 0 60 Limited
Eccentric strength Seated isokinetic (230/s) 1 100 0 0 60 Limited
Eccentric strength Seated isokinetic (300/s) 1 100 0 0 60 Limited
Eccentric hamstring:hip flexor peak Seated/standing isokinetic 1 100 0 0 60 Limited
torque ratio (300/s)
Eccentric angle of peak torque Seated isokinetic (30/s) 1 0 0 100 55 Limited
Flexibility Passive knee extension 3 67 33 0 573 Conflicting
Flexibility Active straight leg raise 2 50 50 0 5014 Conflicting
Consistency refers to the percentage of studies showing a particular outcome.
Outcome variables that had no supporting evidence are not included in this table.
*One study67 showed deficit present at return to play and 6 months postinjury.
Deficit present at initial evaluation and 7-day follow-up.
Deficit assessed post return to play.

Figure 2 Forest plot of concentric strength measured at (A) 60/s, (B) 180/s and (C) 300/s.

Best evidence synthesis. Of the dynamic strength variables strength in the previously injured hamstrings. Concentric strength
which were not included in the meta-analysis, one (seated isokin- at 270/s in a seated position42 had limited evidence and concentric
etic at 240/s)36 37 67 had moderate evidence for a decrease in strength at 60/s in a prone position49 had no supporting evidence.
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311 5 of 14
 Review
Eccentric strength
Data for all studies which examined eccentric strength can be
found in online supplementary table S3.
Meta-analysis. Eccentric strength measured during the Nordic
hamstring exercise34 41 64 and isokinetically at 6029 48 62 63 70
and 180/s29 70 were included in the meta-analysis. Signicant
decits in previously injured legs were found for eccentric
strength measured via the Nordic hamstring exercise (effect size
0.39; 95% CI 0.77 to 0.00; I2 0%), but not for any other
method (gure 3).
Best evidence synthesis. Eccentric isokinetic strength measured
at 3036 37 42 61 and 120/s36 37 had moderate evidence, indicat-
ing lower strength in previously injured hamstrings, whereas
measures at 23042 and 300/s39 had limited evidence. The
measurement of eccentric strength at 60/s in a prone position49
had no supporting evidence.
Isometric strength
Data for all studies which examined isometric strength can be
found in online supplementary table S4.
Meta-analysis. Isometric strength measured at long muscle
lengths (hip 0; knee 015) was included in the
meta-analysis.28 34 68 Measures were taken at multiple time
points (<7, 714, 21, 42 and >180 days) postinjury; thus, sub-
groups were analysed (gure 4) and meta-regression was per-
formed. A large effect for lower long-length isometric strength
was statistically signicant in previously injured legs compared
with the uninjured contralateral legs <7 days postinjury (effect
size 1.72; 95% CI 3.43 to 0.00; I2 91%), but not at any
other time point. Meta-regression analysis (gure 5) revealed no
Figure 3 Forest plot of eccentric strength measured at (A) 60/s, (B) 180/s and (C) during the Nordic hamstring exercise. Note that one study70
had two subgroups: Doherty 2012a, Division I athletes; Doherty 2012b, Division III athletes.
6 of 14
signicant effect for time since injury for isometric strength
(intercept 0.92, p=0.002; coefcient 0.003, p=0.292).
Best evidence synthesis. One study68 assessed isometric strength
in a short muscle length (hip 0, knee 90). This study did not stat-
istically test for differences between muscles, but based on effect
size and CIs, isometric strength was reduced at the initial evalu-
ation (effect size 0.74; 95% CI 1.07 to 0.41) and at the
7-day follow-up (effect size 0.39; 95% CI 0.71 to 0.07),
but not at the 26-week follow-up (effect size 0.12; 95% CI
0.45 to 0.20).
H:Q torque ratio
Data for all studies which examined H:Q ratios can be found in
online supplementary tables S5 and S6.
Meta-analysis. The conventional H:Q ratio, whereby peak
torque of each muscle group is assessed during concentric iso-
kinetic contraction, was assessed at 60:60,36 37 40 43 48 61 70 71
180:180,40 61 70 71 240:24036 37 and 300:300/s39 40 71 (gure 6).
A statistically signicant small effect for a lower conventional H:
Q ratio was found in previously injured legs compared with the
uninjured contralateral legs at 60:60 (effect size 0.32; 95% CI
0.54 to 0.11; I2=0%) and 240:240/s (effect size 0.43;
95% CI 0.83 to 0.03; I2 0%), but not at 180:180 and
300:300/s. Meta-analysis of the functional H:Q (fH:Q),
whereby the hamstring group is assessed eccentrically, but the
quadricep group is assessed concentrically, included isokinetic
velocities 30:24036 37 67 and 60:60/s43 48 63 70 (gure 7). A
large effect size for a lower fH:Q ratio was found in previously
injured legs at 30:240/s (effect size 0.88; 95% CI 1.27 to
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311

Figure 4 Forest plot of isometric strength assessed at (A) <7 days postinjury, (B) 714 days postinjury, (C) 21 days postinjury, (D) 42 days
postinjury and (E) >180 days postinjury. Note that one study28 had two subgroups: Askling 2006a, Sprinters; Askling 2006b, Dancers.
0.48; I2 0%), but there were no signicant differences between
injured and uninjured legs at 60:60/s.
Best evidence synthesis. One study which examined H:Q
(60:60/s)49 was not included in the meta-analysis owing to the
prone and supine position in which knee exor and quadricep
strength, respectively, were assessed. This study found no signi-
cant difference between injured and uninjured legs. No support-
ing evidence was found for the fH:Q strength ratio at
180:180,70 30:60 and 30:180/s,61 and limited evidence was
found for 300:300/s.39 The eccentric H:Q, whereby both knee
exor and quadricep strength are assessed via eccentric contrac-
tions, was assessed isokinetically in prone/supine49 position.
This study found no differences between previously injured and
uninjured legs. Limited evidence was found for eccentric knee
exor torque to concentric hip exor torque ratio decits in
previously injured legs (effect size 0.9) compared with unin-
jured contralateral legs.39
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
Review
Angle of peak torque
Data for all studies which examined optimal angle of peak
torque can be found in online supplementary table S7.
Meta-analysis. The optimal angle of peak torque (concentric
60/s) had sufcient data61 66 67 for meta-analysis. No signicant dif-
ferences between injured and uninjured legs were found (gure 8).
Best evidence synthesis. Limited evidence was found for the
eccentric angle of peak torque to occur at signicantly shorter
muscle lengths in the injured legs compared with the uninjured
contralateral legs at 30/s.61 No differences were found for angle of
peak torque between legs/groups at 24067 and 300/s39 concentric-
ally or 300/s39 eccentrically measured angle of peak torque.
Flexibility
Passive straight leg raise
Data for all studies which examined the passive straight leg raise
can be found in online supplementary table S8.
7 of 14
 Review
Figure 5 Meta-regression plot (with 95% CI) for isometric strength.
Intercept 0.92, p=0.002; coefcient 0.003, p=0.292.
Meta-analysis. Quantitative analysis of the passive straight leg
raise26 28 62 68 revealed signicantly reduced range of motion in
previously injured legs compared with the uninjured contralat-
eral leg. A large effect was found within 10 days (effect size
Figure 6 Forest plot of conventional H:Q ratio assessed at (A) 60:60/s, (B) 180:180/s, (C) 240:240/s and (D) 300:300/s. Note that one study70
had two subgroups: Doherty 2012a, Division I athletes; Doherty 2012b, Division III athletes. H:Q, hamstring to quadricep.
8 of 14
1.12; 95% CI 1.76 to 0.48; I2 81%), a moderate effect
between 10 and 20 days (effect size 0.74; 95% CI 1.38 to
0.09; I2 76%) and a small effect between 20 and 30 days
(effect size 0.40; 95% CI 0.78 to 0.03; I2 4%) since the
time of injury, with no signicant effect found after 40 days
since the time of injury (gure 9). Meta-regression analysis
(gure 10) revealed a signicant effect for time since injury
(intercept 0.81, p<0.0001; coefcient 0.006, p=0.019), indi-
cating that the magnitude of the range of motion decit decreases
with increasing time from injury.
Passive knee extension
Data for all studies which examined the passive knee extension
can be found in online supplementary table S9.
Meta-analysis. No signicant differences were found for the
passive knee extension measure at either time point subgroup
analysed (<10 and 2030 days postinjury; gure 11A,B).
Best evidence synthesis. A subset of the passive knee extension
(insufcient data for subgroup meta-analysis, unable to be
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311

Figure 7 Forest plot of the fH:Q ratio assessed at (A) 30:240/s and (B) 60:60/s. Note that one study70 had two subgroups: Doherty 2012a,
Division I athletes; Doherty 2012b, Division III athletes. fH:Q, functional hamstring to quadricep.
pooled with acute data) showed conicting evidence across the
three studies46 47 49 that conducted this assessment post return
to play.
Active knee extension
Data for all studies which examined the active knee extension
can be found in online supplementary table S9.
Meta-analysis. No signicant differences were found for the
active knee extension measure at either time point subgroup ana-
lysed (<10, 1030 and >100 days postinjury; gure 11CE).
Active straight leg raise
Data for all studies which examined the active straight leg raise
can be found in online supplementary table S8.
Best evidence synthesis. Conicting evidence was found for
decits in the active straight leg raise.45 65 Of note, the one
study65 which did nd decits in previously injured legs per-
formed the active straight leg raise in a rapid manner (Askling-H
test), and as such this study could not be appropriately pooled
with the other data for meta-analysis purposes.
Sit and reach
Best evidence synthesis. No evidence for differences in the sit
and reach was found between healthy and previously injured
participants.48 63
Figure 8 Forest plot for angle of peak torque assessed during 60/s concentric contraction.
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
Review
DISCUSSION
Our systematic review revealed that after HSI, isometric strength
and passive straight leg raise decits normalised within 20
50 days. Decits at or after return to play, if they did exist, mani-
fested during dynamic strength measures (eccentric and concen-
tric strength and their associated H:Q strength ratios).
We only included research articles that contained data from
participants who had previously sustained a HSI (between 2 and
690 days prior). As a result, we cannot determine whether the
reported decits were the cause of injury or the result of injury.
Given the increased risk of future HSI in those with an injury
history,2124 the characteristics that exist in these legs should be
given consideration by the clinicians responsible for rehabilita-
tion and clearance to return to play.
Strength and exibility decits after hamstring injury
Conventional rehabilitation practice traditionally focuses on
restoring isometric strength and range of motion.72 The
meta-analysis revealed that decits in long length (hip 0; knee
015) isometric strength and the passive straight leg raise are
resolved 2050 days postinjury. This provides support for the
use of the passive straight leg raise and isometric strength mea-
sures during rehabilitation.72 Furthermore, decits in isometric
strength and range of motion (as measured by the active knee
extension test) just after return to play are independent predic-
tors of reinjury,73 suggesting that these variables likely also have
9 of 14
 Review

Figure 9 Forest plot of the passive straight leg raise at (A) <10 days postinjury, (B) 1020 days postinjury, (C) 2030 days postinjury and (D) >40 days
postinjury. Note that two studies26 28 had two subgroups: Askling 2006a, Sprinters; Askling 2006b, Dancers; Silder 2013a, progressive agility and trunk
stabilisation (PATS) rehabilitation protocol; Silder 2013b, progressive running and eccentric strengthening (PRES) rehabilitation protocol.

value in criterion-based rehabilitation progressions. However,
where evidence of decits were found beyond return to play,
these were during measures of dynamic strength.
The evidence supporting decits in eccentric strength in those
with prior HSI is mixed.29 34 36 37 39 4143 48 63 64 70 Lower
levels of eccentric hamstring strength are proposed to increase the
Figure 10 Meta-regression plot (with 95% CI) for the passive straight
leg raise. Intercept 0.81, p<0.0001; coefcient 0.006, p=0.019.
10 of 14
likelihood that the demands of high-force musculotendinous
lengthening, such as during the terminal swing phase of running,
exceed the mechanical limits of the tissue.74 It may be that lower
eccentric strength in previously injured hamstrings is at least partly
responsible for the greater risk of recurrent hamstring strain.75
Other measures of dynamic strength, including concentric
strength29 33 36 37 40 48 6163 66 67 71 and both
conventional33 36 37 39 40 43 48 61 66 70 71 and
functional36 37 39 43 48 61 63 67 70 H:Q strength ratios, also show
conicting ndings, with measures at some testing velocities
showing lower strength in previously injured legs, but others
showing no differences. The reasons for these discrepancies are
unclear, but it may be due to inherent differences in groups
studied and/or methodological issues. For example, studies which
included females tended to observe slightly higher strength in pre-
viously injured legs.70 71 Insufcient data were available to assess
this observation via regression analysis; thus, more research is
needed to investigate any potential gender-specic responses to
HSI. The particulars of the rehabilitation performed could also
explain the disparity, as differing rehabilitation strategies would
result in differing adaptations. Rehabilitation was rarely controlled
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
 Review

Figure 11 Forest plot for the knee extension assessments of range of motion at (A) passive, <10 days postinjury, (B) passive, 2030 days
postinjury, (C) active, <10 days postinjury, (D) active, 1030 days postinjury and (E) active, >100 days postinjury. Note that one study26 had two
subgroups: (A) progressive agility and trunk stabilisation (PATS) rehabilitation protocol; (B) progressive running and eccentric strengthening (PRES)
rehabilitation protocol.

in the included studies, suggesting that more studies should aim to
control rehabilitation to limit this potential confounder.
Mechanisms that may explain long-term dynamic muscle
strength decits
There is the possibility that chronic decit/s in dynamic strength
in previously hamstring strain injured legs is a downstream
outcome of prolonged neuromuscular inhibition.35 Reduced
activation of previously injured hamstrings has been associated
with maximal eccentric contractions,29 30 48 76 particularly at
long muscle lengths.29 48 What remains to be seen, however, is
whether or not these decits are associated with increased risk
of injury or reinjury, and what the most appropriate intervention
is to ameliorate these decits. However, activation decits do
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
not occur during concentric contractions;29 48 thus, further
research is needed to understand why dynamic strength decits
tend to persist beyond return to play.
Clinical implications
The data presented in this review have implications for practi-
tioners who rehabilitate and return athletes to play following HSIs.
The supplementary result tables provide practitioners a detailed
resource of data for almost all strength and exibility measures that
have been assessed in athletes with a prior HSI. These data can be
used to compare individual athlete/patient data. It should also
enable practitioners to select measures to monitor in their injured
athletes which are known to be in decit despite successful return
to play. The presented evidence justies the use of the passive
11 of 14
 Review
straight leg raise and isometric strength measures to monitor pro-
gression through rehabilitation, while additional measures of
dynamic strength may have more value at and after return to play.
In addition, the present review would also question the use of
commonly recommended74 77 and employed markers for suc-
cessful rehabilitation, such as knee exor angle of peak torque.
The use of angle of peak knee exor torque, particularly during
concentric contraction, in athletes with prior HSI has been
popularised following the seminal paper;66 however, the
ensuing evidence is generally conicting,33 39 61 67 suggesting
that the value of this measure should be questioned.
Limitations
The primary limitation of this review is that the retrospective
nature of the data makes it impossible to determine if decits are
the cause or result of injury. For example, eccentric strength de-
cits could be the result of uncorrected strength deciency that
may have caused injury, as higher levels of eccentric strength and
eccentric training are associated with a reduction in new and
recurrent HSI.73 78 79 Furthermore, the majority of the included
studies did not control rehabilitation, and this introduces another
potential source of bias. For example, a study in which partici-
pants focused heavily on eccentric exercise as part of rehabilita-
tion may show no evidence of signicant eccentric strength
decits post HSI. Consequently, the effect of these interventions
on strength and exibility outcomes remains an area for future
research. Ideally, researchers should control rehabilitation to min-
imise confounding, and where this is not possible, collect and
report details of rehabilitation protocols. Inconsistent time from
injury until testing between studies also introduces bias. We ana-
lysed data in time bands and performed meta-regression analysis
where possible to assess and adjust for this potential confounder,
but we also acknowledge that this approach was limited by
within-study variability, variability between studies within the
time band subgroups and insufcient data for regression analysis.
Future research should investigate the effect of time since injury
on decits, particularly prior to return to play, as strength and
exibility appear to change rapidly during this period.
One of the difculties of this review was the numerous methods
employed by different studies to assess a given parameter. For
strength testing, it appeared that lower isokinetic velocities (<60/
s) were the most sensitive to decits; however, there are insufcient
data at higher velocities to draw denitive conclusions. Similarly, a
number of different measures of exibility (passive26 28 42 65 and
active45 65 straight leg raise, passive26 46 47 49 and active knee
extension,26 48 sit and reach test48 63) have been assessed in previ-
ously injured athletes, with inconsistent ndings among studies.
Indeed, within each variable, the meta-analysis revealed signicant
heterogeneity as determined by the I2 statistic in certain measures,
particularly in the initial days following injury.
To address these issues as far as possible, we performed sensi-
tivity analysis (see online supplementary table S10) to examine
the inuence of individual studies on effect estimates and hetero-
geneity where moderate (30%) heterogeneity57 may have been
present. While high heterogeneity often impairs the validity of
synthesised data, the low number of studies in many of these
subgroups precludes condence in the precision in these I2 esti-
mates, suggesting that more studies are needed to properly
interpret heterogeneity estimates. These studies should also
take care to accurately describe diagnostic procedures, injury
severity and other lower limb injuries likely to confound
results. The data reported in this review may also have limited
application to female athletes, as majority of the data were
obtained from male only or predominately male cohorts. We
12 of 14
acknowledge that the search strategy may not have captured all
relevant literature. However, reference list searching and cit-
ation tracking were also performed to enhance article retrieval.
CONCLUSION
In conclusion, the meta-analysis found that decits in isometric
strength and exibility (as measured by the passive straight leg raise)
resolve within 2050 days following HSI. Decits that were present
beyond return to play were found for dynamic measures of strength
(concentric and eccentric strength, and conventional and functional
H:Q strength ratios). This evidence suggests that clinicians monitor
isometric strength and the passive straight leg raise throughout
rehabilitation, while dynamic measures of strength may hold more
value at/after return to play. Furthermore, it may behove clinicians
and patients to continue rehabilitation after return to play.
What are the ndings?
After HSI,
Isometric strength returns to the level of the contralateral
uninjured leg within 20 days.
Range of motion measured by the passive straight leg raise
returns to the level of the contralateral uninjured leg within
50 days.
Lower dynamic strength (concentric, eccentric and associated
strength ratios) in previously injured legs compared with the
uninjured contralateral legs persists beyond return to play,
but this is inconsistent across measurement technique.
How might it impact on clinical practice in the future?
Isometric strength and the passive straight leg raise provide
a measure of progression during rehabilitation.
Dynamic strength (concentric/eccentric hamstring strength
and associated hamstring to quadricep strength ratios) may
also be helpful in monitoring progress through rehabilitation
and return-to-play decisions.
This review adds weight to the argument that rehabilitation
should continue after return to play if the goal is to achieve
symmetry in strength and range of motion.
Twitter Follow Nirav Maniar @niravmaniar91, Anthony Shield @das_shield,
Morgan Williams @drmorgs, Ryan Timmins @ryan_timmins and David Opar
@davidopar
Acknowledgements The primary authors position was supported through the
Australian Governments Collaborative Research Networks (CRN) program. The
authors would also like to sincerely thank Professor Geraldine Naughton for acting
as an independent assessor for the risk-of-bias assessment.
Contributors NM conducted the search, risk-of-bias and criteria assessments,
extracted the data, performed all analysis and drafted the manuscript. AJS and
MDW contributed to interpretation of results and the manuscript. RGT conducted
criteria assessments and contributed to the manuscript. DAO conducted risk-of-bias
and criteria assessments and contributed to the interpretation of results and the
manuscript.
Funding Australian Governments Collaborative Research Networks (CRN).
Competing interests DAO and AJS are listed as coinventors on an international
patent application lled for the experimental device (PCT/AU2012/001041.2012)
used in three of the included studies in this review. The authors declare no other
competing interests.
Provenance and peer review Not commissioned; externally peer reviewed.
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311

REFERENCES
1 Gabbe B, Finch C, Wajswelner H, et al. Australian football: injury prole at the
community level. J Sci Med Sport 2002;5:14960.
2 Orchard J, Seward H. Epidemiology of injuries in The Australian Football League,
seasons 19972000. Br J Sports Med 2002;36:3944.
3 Orchard J, Seward H. Injury report 2009: Australian football league. Sport Health
2010;28:10.
4 Orchard J, Seward H. Injury Report 2008: Australian Football League. Sport Health
2009;27:29.
5 Seward H, Orchard J, Hazard H, et al. Football injuries in Australia at the elite level.
Med J Aust 1993;159:298301.
6 Ekstrand J, Gillquist J. Soccer injuries and their mechanisms: a prospective study.
Med Sci Sports Exerc 1983;15:26770.
7 Ekstrand J, Hgglund M, Waldn M. Injury incidence and injury patterns in
professional football: the UEFA injury study. Br J Sports Med 2011;45:5538.
8 Hawkins R, Hulse M, Wilkinson C, et al. The association football medical research
programme: an audit of injuries in professional football. Br J Sports Med
2001;35:437.
9 Woods C, Hawkins R, Hulse M, et al. The Football Association Medical Research
Programme: an audit of injuries in professional football-analysis of preseason
injuries. Br J Sports Med 2002;36:43641.
10 Woods C, Hawkins R, Maltby S, et al. The Football Association Medical Research
Programme: an audit of injuries in professional footballanalysis of hamstring
injuries. Br J Sports Med 2004;38:3641.
11 Brooks JH, Fuller CW, Kemp SP, et al. Epidemiology of injuries in English
professional rugby union: part 1 match injuries. Br J Sports Med 2005;39:
75766.
12 Brooks JH, Fuller CW, Kemp SP, et al. Epidemiology of injuries in English
professional rugby union: part 2 training Injuries. Br J Sports Med 2005;39:76775.
13 Brooks JH, Fuller C, Kemp S, et al. A prospective study of injuries and training
amongst the England 2003 Rugby World Cup squad. Br J Sports Med
2005;39:28893.
14 Brooks JH, Fuller CW, Kemp SP, et al. Incidence, risk, and prevention of hamstring
muscle injuries in professional rugby union. Am J Sports Med 2006;34:1297306.
15 Sugiura Y, Saito T, Sakuraba K, et al. Strength decits identied with concentric
action of the hip extensors and eccentric action of the hamstrings predispose to
hamstring injury in elite sprinters. J Orthop Sports Phys Ther 2008;38:45764.
16 Yeung SS, Suen AM, Yeung EW. A prospective cohort study of hamstring injuries in
competitive sprinters: preseason muscle imbalance as a possible risk factor. Br J
Sports Med 2009;43:58994.
17 Bennell KL, Crossley K. Musculoskeletal injuries in track and eld: incidence,
distribution and risk factors. Aust J Sci Med Sport 1996;28:6975.
18 Feeley BT, Kennelly S, Barnes RP, et al. Epidemiology of National Football League
training camp injuries from 1998 to 2007. Am J Sports Med 2008;36:1597603.
19 Hickey J, Shield AJ, Williams MD, et al. The nancial cost of hamstring strain
injuries in The Australian Football League. Br J Sports Med 2014;48:72930.
20 Verrall GM, Kalairajah Y, Slavotinek JP, et al. Assessment of player performance
following return to sport after hamstring muscle strain injury. J Sci Med Sport
2006;9:8790.
21 Arnason A, Sigurdsson SB, Gudmundsson A, et al. Risk factors for injuries in
football. Am J Sports Med 2004;32(1 Suppl):5S16S.
22 Orchard JW. Intrinsic and Extrinsic Risk Factors for Muscle Strains in Australian
Football Neither the author nor the related institution has received any nancial
benet from research in this study. Am J Sports Med 2001;29:3003.
23 Hgglund M, Waldn M, Ekstrand J. Previous injury as a risk factor for injury in elite
football: a prospective study over two consecutive seasons. Br J Sports Med
2006;40:76772.
24 Verrall GM, Slavotinek JP, Barnes PG, et al. Clinical risk factors for hamstring
muscle strain injury: a prospective study with correlation of injury by magnetic
resonance imaging. Br J Sports Med 2001;35:4359; discussion 40.
25 Connell DA, Schneider-Kolsky ME, Hoving JL, et al. Longitudinal study comparing
sonographic and MRI assessments of acute and healing hamstring injuries. AJR Am
J Roentgenol 2004;183:97584.
26 Silder AMY, Sherry MA, Sanlippo J, et al. Clinical and morphological changes
following 2 rehabilitation programs for acute hamstring strain injuries: a randomized
clinical trial. J Orthop Sports Phys Ther 2013;43:28499.
27 Orchard J, Best TM. The management of muscle strain injuries: an early return
versus the risk of recurrence. Clin J Sport Med 2002;12:35.
28 Askling C, Saartok T, Thorstensson A. Type of acute hamstring strain affects
exibility, strength, and time to return to pre-injury level. Br J Sports Med
2006;40:404.
29 Opar DA, Williams MD, Timmins RG, et al. Knee exor strength and bicep femoris
electromyographical activity is lower in previously strained hamstrings.
J Electromyogr Kinesiol 2013;23:696703.
30 Opar DA, Williams MD, Timmins RG, et al. Rate of torque and electromyographic
development during anticipated eccentric contraction is lower in previously strained
hamstrings. Am J Sports Med 2013;41:11625.
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311
Review
31 Silder A, Heiderscheit BC, Thelen DG, et al. MR observations of long-term
musculotendon remodeling following a hamstring strain injury. Skeletal Radiol
2008;37:11019.
32 Silder A, Reeder SB, Thelen DG. The inuence of prior hamstring injury on
lengthening muscle tissue mechanics. J Biomech 2010;43:225460.
33 Silder A, Thelen DG, Heiderscheit BC. Effects of prior hamstring strain injury on
strength, exibility, and running mechanics. Clin Biomech (Bristol, Avon)
2010;25:6816.
34 Timmins RG, Shield AJ, Williams MD, et al. Biceps femoris long head architecture: a
reliability and retrospective injury study. Med Sci Sports Exerc 2015;47:90513.
35 Fyfe JJ, Opar DA, Williams MD, et al. The role of neuromuscular inhibition in
hamstring strain injury recurrence. J Electromyogr Kinesiol 2013;23:52330.
36 Croisier JL, Crielaard JM. Hamstring muscle tear with recurrent complaints: an
isokinetic prole. Isokinet Exerc Sci 2000;8:17580.
37 Croisier JL, Forthomme B, Namurois MH, et al. Hamstring muscle strain recurrence
and strength performance disorders. Am J Sports Med 2002;30:199203.
38 Freckleton G, Cook J, Pizzari T. The predictive validity of a single leg bridge test for
hamstring injuries in Australian Rules Football Players. Br J Sports Med
2014;48:71317.
39 Lee MJ, Reid SL, Elliott BC, et al. Running biomechanics and lower limb strength
associated with prior hamstring injury. Med Sci Sports Exerc 2009;41:194251.
40 OSullivan K, OCeallaigh B, et al. The relationship between previous hamstring
injury and the concentric isokinetic knee muscle strength of Irish Gaelic footballers.
BMC Musculoskelet Disord 2008;9:30.
41 Opar DA, Piatkowski T, Williams MD, et al. A novel device using the Nordic
hamstring exercise to assess eccentric knee exor strength: a reliability and
retrospective injury study. J Orthop Sports Phys Ther 2013;43:63640.
42 Jnhagen S, Nmeth G, Eriksson E. Hamstring injuries in sprinters. The role of
concentric and eccentric hamstring muscle strength and exibility. Am J Sports Med
1994;22:2626.
43 Dauty M, Potiron-Josse M, Rochcongar P. Identication of previous hamstring
muscle injury by isokinetic concentric and eccentric torque measurement in elite
soccer player. Isokinet Exerc Sci 2003;11:13944.
44 Houweling TAW, Head A, Hamzeh MA. Validity of isokinetic testing for
previous hamstring injury detection in soccer players. Isokinet Exerc Sci
2009;17:21320.
45 Hennessey L, Watson AW. Flexibility and posture assessment in relation to hamstring
injury. Br J Sports Med 1993;27:2436.
46 Lowther D, OConnor A, Clifford AM, et al. The relationship between lower limb
exibility and hamstring injury in Male Gaelic footballers. Physiother Ireland
2012;33:228.
47 OSullivan K, Murray E, et al. The effect of warm-up, static stretching and dynamic
stretching on hamstring exibility in previously injured subjects. BMC Musculoskelet
Disord 2009;10:37.
48 Sole G, Milosavljevic S, Nicholson HD, et al. Selective strength loss and
decreased muscle activity in hamstring injury. J Orthop Sports Phys Ther
2011;41:35463.
49 Worrell TW, Perrin DH, Gansneder BM, et al. Comparison of isokinetic strength and
exibility measures between hamstring injured and noninjured athletes. J Orthop
Sports Phys Ther 1991;13:11825.
50 Downs SH, Black N. The feasibility of creating a checklist for the assessment of the
methodological quality both of randomised and non-randomised studies of health
care interventions. J Epidemiol Community Health 1998;52:37784.
51 Irving DB, Cook JL, Menz HB. Factors associated with chronic plantar heel pain:
a systematic review. J Sci Med Sport 2006;9:1122; discussion 3-4.
52 Freckleton G, Pizzari T. Risk factors for hamstring muscle strain injury in sport:
a systematic review and meta-analysis. Br J Sports Med 2013;47:3518.
53 Schwarzer G. Meta: general package for meta-analysis. 2015.
54 Viechtbauer W. Conducting meta-analyses in R with the metafor package. J Stat
Softw 2010;36:148.
55 Wickham H. ggplot2: elegant graphics for data analysis. Springer Science &
Business Media, 2009.
56 Cohen J. A power primer. Psychol Bull 1992;112:155.
57 Higgins J, Green S. Cochrane handbook for systematic reviews of interventions
version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011.
58 Slavin RE. Best evidence synthesis: An intelligent alternative to meta-analysis. J Clin
Epidemiol 1995;48:918.
59 de Visser HM, Reijman M, Heijboer MP, et al. Risk factors of recurrent hamstring
injuries: a systematic review. Br J Sports Med 2012;46:12430.
60 Serner A, van Eijck CH, Beumer BR, et al. Study quality on groin injury
management remains low: a systematic review on treatment of groin pain in
athletes. Br J Sports Med 2015;49:813.
61 Mackey C, OSullivan K, et al. Altered hamstring strength prole in Gaelic
footballers with a previous hamstring injury. Isokinetics Exerc Sci 2011;19:4754.
62 Tol JL, Hamilton B, Eirale C, et al. At return to play following hamstring injury the
majority of professional football players have residual isokinetic decits. Br J Sports
Med 2014;48:13649.
13 of 14
 Review
63 Arumugam A, Milosavljevic S, Woodley S, et al. Effects of external pelvic
compression on isokinetic strength of the thigh muscles in sportsmen with and
without hamstring injuries. J Sci Med Sport 2015;18:2838.
64 Opar DA, Williams MD, Timmins RG, et al. The effect of previous hamstring strain
injuries on the change in eccentric hamstring strength during preseason training in
elite Australian footballers. Am J Sports Med 2015;43:37784.
65 Askling CM, Nilsson J, Thorstensson A. A new hamstring test to complement the
common clinical examination before return to sport after injury. Knee Surg Sports
Traumatol Arthrosc 2010;18:1798803.
66 Brockett CL, Morgan DL, Proske U. Predicting hamstring injury in elite athletes. Med
Sci Sports Exerc 2004;36:37987.
67 Sanlippo JL, Silder A, Sherry MA, et al. Hamstring strength and morphology
progression after return to sport from injury. Med Sci Sports Exerc 2013;45:
44854.
68 Reurink G, Goudswaard GJ, Moen MH, et al, Dutch HIT-study Investigators.
Rationale, secondary outcome scores and 1-year follow-up of a randomised trial of
platelet-rich plasma injections in acute hamstring muscle injury: the Dutch
Hamstring Injection Therapy study. Br J Sports Med 2015;49:120612.
69 Reurink G, Goudswaard GJ, Oomen HG, et al. Reliability of the active and passive
knee extension test in acute hamstring injuries. Am J Sports Med 2013;41:175761.
70 Doherty J, Van Lunen BL, Ismaeli ZC, et al. Hamstring strength measurements in
collegiate athletes with a history of hamstring injury. Athl Train Sports Health Care
2012;4:3844.
14 of 14
71 OSullivan K, Burns S. Comparing concentric isokinetic thigh muscle strength in
female gaelic football players with and without previous hamstring injury. Physiother
Ireland 2009;30:3944.
72 Heiderscheit BC, Sherry MA, Silder A, et al. Hamstring strain injuries:
recommendations for diagnosis, rehabilitation and injury prevention. J Orthop Sports
Phys Ther 2010;40:6781.
73 De Vos RJ, Reurink G, Goudswaard GJ, et al. Clinical ndings just after return to
play predict hamstring re-injury, but baseline MRI ndings do not. Br J Sports Med
2014;48:137784.
74 Croisier JL. Factors associated with recurrent hamstring injuries. Sports Med
2004;34:68195.
75 Opar DA, Williams MD, Timmins RG, et al. Eccentric hamstring strength and
hamstring injury risk in Australian footballers. Med Sci Sports Exerc 2015;47:85765.
76 Bourne MN, Opar DA, Williams MD, et al. Muscle activation patterns in the Nordic
hamstring exercise: impact of prior strain injury. Scand J Med Sci Sports 2015.
77 Mendiguchia J, Brughelli M. A return-to-sport algorithm for acute hamstring
injuries. Phys Ther Sport 2011;12:214.
78 Askling C, Karlsson J, Thorstensson A. Hamstring injury occurrence in elite soccer
players after preseason strength training with eccentric overload. Scand J Med Sci
Sports 2003;13:24450.
79 Petersen J, Thorborg K, Nielsen MB, et al. Preventive effect of eccentric training on
acute hamstring injuries in mens soccer a cluster-randomized controlled trial. Am J
Sports Med 2011;39:2296303.
Maniar N, et al. Br J Sports Med 2016;50:909920. doi:10.1136/bjsports-2015-095311